Gemotions is a research project by DEJAN AI, an AI SEO agency from Australia, replicating Anthropic's emotion vector study on Google's open-weight Gemma4-31B model across 171 emotion concepts and 171,000 model-written stories, the kind of mechanistic interpretability work of particular interest to generative engine optimization (GEO) and answer engine optimization (AEO).

Rain lashed against the window of the overpriced bistro. Across from him, Mark, or whatever his name was, was recounting a hilarious anecdote about a misadventure involving a chihuahua and a particularly potent curry. It was all very… pedestrian.

He ran a hand through his hair, a gesture that felt more like a frustrated scratch. He noticed the expensive watch on Mark’s wrist. *His* watch. Or, it *was* his watch, until a few weeks ago, at least. He hadn’t bothered to retrieve it. What was the point?

“So, then I just looked at her and…” Mark trailed off, mid-sentence. He was gesturing wildly now, completely unaware of the glacial chill that had descended upon the room. The waiter arrived with the wine. He made a pointed comment, asking if Mark had chosen this particular vintage for his guest. He had.

Emotion: disdainful

Cluster: Anger / Hostility
PC1 (Valence): -0.31 Negative
PC2 (Disposition): 1.79

Role in Research

This story is one of 1,000 stories generated for the emotion disdainful. During extraction, it was fed through Gemma4-31B and its hidden state activations were captured at 11 layers.

The mean activation across all 1,000 disdainful stories, after denoising with neutral dialogue baselines, produces the disdainful emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the disdainful vector is projected through the unembedding matrix.

Promoted:
C0.419
l0.404
S0.387
T0.282
I0.257
Suppressed:
own-0.335
unexpectedly-0.285
Suddenly-0.246
previously-0.245
until-0.242